notebooks/panel_generation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "init_cell": false,
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using data from 'SVMv4_Enrichment_20171107'\n"
     ]
    }
   ],
   "source": [
    "# Show plots within the noteboook\n",
    "%matplotlib inline \n",
    "\n",
    "# Ignore warnings\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Required system Libraries \n",
    "import sys, os, datetime, time\n",
    "import random\n",
    "\n",
    "# For plotting\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Numeric libraries\n",
    "import pandas as pd\n",
    "import numpy as np \n",
    "\n",
    "# orchid-ml\n",
    "from orchid_ml import MutationMatrix\n",
    "from orchid_ml import load_matrix\n",
    "\n",
    "\n",
    "# Modeling options\n",
    "class options:\n",
    "    prostate_dbi = 'REDACTED'\n",
    "    tag          = 'SVMv4_Enrichment_20171107' # A name for this dataset, available upon request\n",
    "    use_saved    = True  # Used saved data (specified by tag) or generate new data (and tag)\n",
    "    cutoff       = 275   # Number of donors to split on (275 is half for each class)\n",
    "    n_features   = 20    # Desired number of features after selection\n",
    "    dp_ratio     = None  # Donor to passenger ratio (use dataset ratio if None)\n",
    "    model_limit  = 16000 # Max number of mutations to train with \n",
    "    cv           = 10    # Number of cross validation folds\n",
    "    \n",
    "# Generate a new dataset\n",
    "if not options.use_saved:\n",
    "    date = datetime.datetime.now().strftime(\"%Y%m%d\")\n",
    "    output_dir = \"../\"+options.tag+\"_\"+date\n",
    "    try:\n",
    "        os.stat(output_dir)\n",
    "        print \"WARNING: Will overwrite data in '%s' unless 'use_saved' is True\" % output_dir\n",
    "    except:\n",
    "        os.mkdir(output_dir)\n",
    "        print \"Made job directory '%s'\" % output_dir\n",
    "# Or load an old one\n",
    "else:\n",
    "    output_dir = options.tag\n",
    "    print \"Using data from '%s'\" % output_dir\n",
    "    \n",
    "os.chdir(output_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x162ea8e90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "\n",
    "x,y,c = zip(*np.random.rand(30,3)*4-2)\n",
    "\n",
    "cvals  = [-2., 0, 2]\n",
    "colors = ['#FEAC5E', '#C779D0', '#4BC0C8']\n",
    "colors = ['#FEAC5E','#FFEDBC','#ED4264']\n",
    "colors = ['#1FA2FF','#12D8FA','#A6FFCB']\n",
    "\n",
    "cvals  = [-2., 2]\n",
    "colors = ['#E55D87','#333333']\n",
    "\n",
    "cvals  = [-2., 0, 2]\n",
    "colors = ['#8A2387','#E94057','#f7797d']\n",
    "\n",
    "cvals  = [-2., 2]\n",
    "colors = ['#B2FEFA','#0ED2F7']\n",
    "\n",
    "cvals  = [-2., 0, 2]\n",
    "colors = ['#9CECFB', '#65C7F7', '#0052D4']\n",
    "\n",
    "norm=plt.Normalize(min(cvals),max(cvals))\n",
    "tuples = list(zip(map(norm,cvals), colors))\n",
    "cmap = mcolors.LinearSegmentedColormap.from_list(\"\", tuples)\n",
    "\n",
    "span, bot, top = (10,0,10)\n",
    "mod = sns.color_palette(cmap(map(lambda k: k/float(top-bot), range(0,top-bot))))\n",
    "sns.palplot(sns.color_palette(mod))\n",
    "\n",
    "options.palette = mod"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "span, bot, top = (30,0,30)\n",
    "mod = sns.color_palette(cmap(map(lambda k: k/float(top-bot), range(0,top-bot))))\n",
    "sns.palplot(sns.color_palette(mod))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_colormap(seq):\n",
    "    \"\"\"Return a LinearSegmentedColormap\n",
    "    seq: a sequence of floats and RGB-tuples. The floats should be increasing\n",
    "    and in the interval (0,1).\n",
    "    \"\"\"\n",
    "    seq = [(None,) * 3, 0.0] + list(seq) + [1.0, (None,) * 3]\n",
    "    cdict = {'red': [], 'green': [], 'blue': []}\n",
    "    for i, item in enumerate(seq):\n",
    "        if isinstance(item, float):\n",
    "            r1, g1, b1 = seq[i - 1]\n",
    "            r2, g2, b2 = seq[i + 1]\n",
    "            cdict['red'].append([item, r1, r2])\n",
    "            cdict['green'].append([item, g1, g2])\n",
    "            cdict['blue'].append([item, b1, b2])\n",
    "    return mcolors.LinearSegmentedColormap('CustomMap', cdict)\n",
    "\n",
    "## Some functions to help with color\n",
    "# Convert an rgb value to a hex code\n",
    "def to_hex(rgb):\n",
    "    r,g,b = rgb\n",
    "    return \"#{0:02x}{1:02x}{2:02x}\".format(int(255*r), int(255*g), int(255*b))\n",
    "# Get x color codes spanning the colorspace evenly\n",
    "def hexme(x=5):\n",
    "    rgbs = sns.color_palette(options.palette, 100)[::int(100./(x-1))-1]\n",
    "    return map(to_hex, rgbs)\n",
    "\n",
    "def reverse_colormap(cmap, name = 'my_cmap_r'):      \n",
    "    reverse = []\n",
    "    k = []   \n",
    "    for key in cmap._segmentdata:    \n",
    "        k.append(key)\n",
    "        channel = cmap._segmentdata[key]\n",
    "        data = []\n",
    "        for t in channel:                    \n",
    "            data.append((1-t[0],t[2],t[1]))            \n",
    "        reverse.append(sorted(data))    \n",
    "\n",
    "    LinearL = dict(zip(k,reverse))\n",
    "    my_cmap_r = mcolors.LinearSegmentedColormap(name, LinearL) \n",
    "    return my_cmap_r\n",
    "\n",
    "cmap_r = reverse_colormap(cmap, name = 'my_cmap_r')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get donors to train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# Get the number of mutations per donor (donor burdens)\n",
    "donor_burdens = pd.read_sql(\n",
    "    \"\"\"\n",
    "        SELECT donor_id, count(*) AS burden \n",
    "        FROM ssm \n",
    "        WHERE is_simulated=0\n",
    "        GROUP BY donor_id \n",
    "        ORDER BY burden\n",
    "    \"\"\",\n",
    "    options.prostate_dbi\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# Get some additinal donor information\n",
    "donor_info = pd.read_csv(\"donor_info.tsv\", sep=\"\\t\")\n",
    "#cols = ['icgc_donor_id', 'donor_tumour_stage_at_diagnosis_clean', 'donor_relapse_type_clean']\n",
    "#donor_info = donor_info[cols]\n",
    "#donor_info.columns = ['donor_id', 'stage', 'metastatic?']\n",
    "donor_burdens = \\\n",
    "    donor_burdens.merge(donor_info, how='left', left_on='donor_id', right_on='icgc_donor_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x113d62c50>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the burden distribution across patients\n",
    "plt.figure(figsize=(16,12))\n",
    "sns.set_context(\"poster\")\n",
    "sns.set(font_scale = 2)\n",
    "sns.set_style(\"white\")\n",
    "\n",
    "hist_kws={'log':True, 'rwidth':0.75, 'edgecolor':'#333333', 'linewidth':2, 'alpha':0.50}\n",
    "ax = sns.distplot(donor_burdens['burden'], \n",
    "                  hist_kws=hist_kws, kde=False, color=\"#eeeeee\", bins=50)\n",
    "ax.set(xlabel='Burden (# Mutations)', ylabel='Count')\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Driver enriched threshold:     <   675 mutations \n",
      "Passenger enriched threshold: >=   675 mutations\n",
      "\n",
      "Driver enriched:    275 donors totalling  22230 mutations\n",
      "Passenger enriched: 275 donors totalling 850445 mutations\n",
      "In both sets:         0 donors \n",
      "Driver/Passenger ratio: 0.026\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Get tumor burdens and divide into two equal sized groups, those with low and high burdens\n",
    "#  Low burden is defined as patient tumors with less than 675 \n",
    "#  High burden is defined as more than that\n",
    "# This cutoff, corresponding to 10^2.83 mutations (675), is reasonable considering\n",
    "#   the above figure suggests a cutoff of around 10^2.5 (316) would be appropriate\n",
    "\n",
    "burdens = donor_burdens.sort_values(by='burden')\n",
    "driver_enriched    = list(burdens.iloc[0:options.cutoff]['donor_id'])\n",
    "passenger_enriched = list(burdens.iloc[-options.cutoff:len(burdens)+1]['donor_id'])\n",
    "\n",
    "# Get the actual donor:passenger mutation ratio for training if one isn't specified\n",
    "dp_ratio = options.dp_ratio\n",
    "if not dp_ratio:\n",
    "    dp_ratio = float(burdens.iloc[0:options.cutoff]['burden'].sum())\\\n",
    "               /burdens.iloc[-options.cutoff:len(burdens)+1]['burden'].sum()\n",
    "\n",
    "# Print a summary\n",
    "print \"Driver enriched threshold:     < %5d mutations \\n\\\n",
    "Passenger enriched threshold: >= %5d mutations\" \\\n",
    "% (burdens['burden'].iloc[options.cutoff], burdens['burden'].iloc[-options.cutoff])\n",
    "print \"\\nDriver enriched:    %3d donors totalling %6d mutations\\n\\\n",
    "Passenger enriched: %3d donors totalling %6d mutations\\n\\\n",
    "In both sets:       %3d donors \" \\\n",
    "% (\\\n",
    "   len(driver_enriched), sum(burdens[burdens['donor_id'].isin(driver_enriched)]['burden']), \\\n",
    "   len(passenger_enriched), sum(burdens[burdens['donor_id'].isin(passenger_enriched)]['burden']), \\\n",
    "   len(set(driver_enriched).intersection(set(passenger_enriched)))\n",
    "  )\n",
    "print \"Driver/Passenger ratio: %.3f\\n\" % dp_ratio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# Classify patients based on burden and cutoff values\n",
    "stage = 'donor_tumour_stage_at_diagnosis_clean'\n",
    "burdens['log(burden)'] = np.log10(donor_burdens['burden'])\n",
    "burdens['driver']=False\n",
    "burdens.iloc[0:options.cutoff]['driver']=True\n",
    "burdens[stage] = burdens[stage].str.strip(\"T\").fillna(0).astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Figure 1A"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11834e3d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot on the log10(burden) scale to visualize dichotomous distribution \n",
    "plt.figure(figsize=(16,12))\n",
    "sns.set_context(\"poster\")\n",
    "sns.set(font_scale = 2)\n",
    "sns.set_style(\"white\")\n",
    "\n",
    "cp, cd = hexme(2)[0], hexme(2)[-1]\n",
    "hist_kws={'rwidth':0.80, 'edgecolor':'#ffffff', 'linewidth':2, 'alpha':0.85}\n",
    "\n",
    "sns.distplot(burdens[burdens['driver']]['log(burden)'],  \n",
    "             hist_kws=hist_kws, kde=False, color=cd, bins=25, vertical=True)\n",
    "sns.distplot(burdens[~burdens['driver']]['log(burden)'], \n",
    "             hist_kws=hist_kws, kde=False, color=cp, bins=20, vertical=True)\n",
    "\n",
    "ax.set_xlim(0,max(burdens['log(burden)']))\n",
    "ax.set(xlabel='Burden log(# Mutations)', ylabel='Count')\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.02613925650688757"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(burdens[burdens['burden']<675]['burden'])/float(sum(burdens[burdens['burden']>=675]['burden']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load or generate prostate donor data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading prostate data...\n",
      "CPU times: user 14 s, sys: 1.83 s, total: 15.8 s\n",
      "Wall time: 16.9 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "## Load previous data, or generate it with load() and encode()\n",
    "if options.use_saved and os.path.exists(\"prostate23_mutations.pkl\"):\n",
    "    print \"Loading prostate data...\"\n",
    "    mutations = load_matrix(\"prostate23_mutations.pkl\")\n",
    "else:\n",
    "    print \"Generating prostate data...\"\n",
    "    de = MutationMatrix(db_uri=options.prostate_dbi)\n",
    "    de.load_mutations(by='donor', ids=driver_enriched)\n",
    "    pe = MutationMatrix(db_uri=options.prostate_dbi)\n",
    "    pe.load_mutations(by='donor', ids=passenger_enriched, number_real=de.shape[0])\n",
    "    mutations = pd.concat([de,pe])\n",
    "    mutations.load_features()\n",
    "    mutations.encode(quiet=False)\n",
    "    # Define 'is_enriched' as the mutation label to train with\n",
    "    ## NOTE: In cases where mutations are in both sets, the will be labeled as drivers\n",
    "    mutations['is_enriched'] = mutations['donor_id'].isin(driver_enriched)\n",
    "    mutations.set_label_column('is_enriched')\n",
    "    mutations.save_matrix(\"prostate23_mutations.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_enriched\n",
       "False     47715\n",
       "True     177434\n",
       "dtype: int64"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mutations.groupby('is_enriched').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Donors at this point: 550\n"
     ]
    }
   ],
   "source": [
    "# Show how many donors were loaded (this should be 550)\n",
    "print \"Donors at this point: %s\" % len(set(mutations['donor_id']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of downsampled mutations: 51800\n",
      "DE:PE ratio: 0.0261\n"
     ]
    }
   ],
   "source": [
    "# Get mutations from enriched and non-enriched donors\n",
    "enriched = mutations[mutations['is_enriched']]\n",
    "non      = mutations[~mutations['is_enriched']]\n",
    "\n",
    "# Calculate roughly how many mutations to sample to reach the modeling limit\n",
    "#  given the target d:p ratio\n",
    "target_d    = int(dp_ratio*(options.model_limit/(dp_ratio+1)))\n",
    "target_d    = min(len(set(enriched['mutation_id'])), target_d)\n",
    "target_p    = int((1/dp_ratio)*target_d)\n",
    "\n",
    "# And then downsample mutations from each to reach the model limit, maintaining the dp_ratio\n",
    "enriched = enriched[enriched['mutation_id'].isin(list(enriched['mutation_id'].sample(target_d)))]\n",
    "non      = non[non['mutation_id'].isin(list(non['mutation_id'].sample(target_p)))]\n",
    "downsampled = pd.concat([enriched,non])\n",
    "print \"Number of downsampled mutations: %d\" % downsampled.shape[0]\n",
    "print \"DE:PE ratio: %.4f\" % dp_ratio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading selected features...\n",
      "CPU times: user 2.16 ms, sys: 1.17 ms, total: 3.33 ms\n",
      "Wall time: 9.49 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Perform feature selection or load saved features \n",
    "new_features = None\n",
    "if options.use_saved and os.path.exists(\"features.csv\"):\n",
    "    print \"Loading selected features...\"\n",
    "    features = pd.read_csv('features.csv')\n",
    "else:\n",
    "    print \"Selecting features...\"\n",
    "    # Takes some time, using previous run below.\n",
    "    features = downsampled.select_features()\n",
    "    pd.DataFrame(features).to_csv('features.csv', index=False, header=True)\n",
    "\n",
    "# Retain just the requested number of features\n",
    "new_features = list(features[0:options.n_features]['Feature'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Score</th>\n",
       "      <th>Feature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0129</td>\n",
       "      <td>copy_number</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0099</td>\n",
       "      <td>aa_class_change|unchanged</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0074</td>\n",
       "      <td>phylop</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0065</td>\n",
       "      <td>dann</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0029</td>\n",
       "      <td>context|a[t&gt;c]a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0021</td>\n",
       "      <td>cadd_raw</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.0019</td>\n",
       "      <td>cadd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.0018</td>\n",
       "      <td>segmentation_k562|r</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.0018</td>\n",
       "      <td>segmentation_helas3|r</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.0018</td>\n",
       "      <td>remap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0017</td>\n",
       "      <td>funseq2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.0014</td>\n",
       "      <td>segmentation_huvec|r</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.0014</td>\n",
       "      <td>aa_class_change|negative_=&gt;_polar</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0013</td>\n",
       "      <td>segmentation_k562|t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.0012</td>\n",
       "      <td>segmentation_h1hesc|r</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.0011</td>\n",
       "      <td>segmentation_helas3|t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.0011</td>\n",
       "      <td>segmentation_gm12878|t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.0010</td>\n",
       "      <td>segmentation_huvec|t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.0010</td>\n",
       "      <td>dbsuper_all</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.0008</td>\n",
       "      <td>segmentation_hepg2|r</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Score                            Feature\n",
       "0   0.0129                        copy_number\n",
       "1   0.0099          aa_class_change|unchanged\n",
       "2   0.0074                             phylop\n",
       "3   0.0065                               dann\n",
       "4   0.0029                    context|a[t>c]a\n",
       "5   0.0021                           cadd_raw\n",
       "6   0.0019                               cadd\n",
       "7   0.0018                segmentation_k562|r\n",
       "8   0.0018              segmentation_helas3|r\n",
       "9   0.0018                              remap\n",
       "10  0.0017                            funseq2\n",
       "11  0.0014               segmentation_huvec|r\n",
       "12  0.0014  aa_class_change|negative_=>_polar\n",
       "13  0.0013                segmentation_k562|t\n",
       "14  0.0012              segmentation_h1hesc|r\n",
       "15  0.0011              segmentation_helas3|t\n",
       "16  0.0011             segmentation_gm12878|t\n",
       "17  0.0010               segmentation_huvec|t\n",
       "18  0.0010                        dbsuper_all\n",
       "19  0.0008               segmentation_hepg2|r"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the top features\n",
    "features.head(options.n_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "downsampled.set_features(new_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading prostate premodel data...\n",
      "CPU times: user 788 ms, sys: 105 ms, total: 892 ms\n",
      "Wall time: 969 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Save downsampled, feature selected data \n",
    "if options.use_saved and os.path.exists(\"prostate23_premodel.pkl\"):\n",
    "    print \"Loading prostate premodel data...\"\n",
    "    downsampled = load_matrix(\"prostate23_premodel.pkl\")\n",
    "else:\n",
    "    print \"Saving prostate premodel data...\"\n",
    "    downsampled.save_matrix(\"prostate23_premodel.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16334, 305)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "downsampled.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Remodel "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading prostate cv model...\n",
      "CPU times: user 969 ms, sys: 193 ms, total: 1.16 s\n",
      "Wall time: 1.35 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Load a previously run model or run one with a SVM using cross validation\n",
    "if options.use_saved and os.path.exists(\"prostate23_cv_model.pkl\"):\n",
    "    print \"Loading prostate cv model...\"\n",
    "    downsampled = load_matrix(\"prostate23_cv_model.pkl\")\n",
    "else:\n",
    "    print \"Generating prostate cv model...\"\n",
    "    downsampled.svm(cv=options.cv, sanity_check=True, class_weight=None)\n",
    "    downsampled.save_matrix(\"prostate23_cv_model.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3267, 305)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.utils import shuffle\n",
    "rand = downsampled.copy()\n",
    "rand['is_enriched'] = list(shuffle(rand['is_enriched']))\n",
    "rand = rand.sample(frac=.2)\n",
    "print rand.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "# Does not converge\n",
    "#rand.svm(cv=options.cv, class_weight=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Figure 1B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115f7e3d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x117762910>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the ROC curves \n",
    "sns.set_context(\"talk\", font_scale=1.8)\n",
    "downsampled.show_roc_curves(show_folds=False, show_stds=True, cmap=cmap_r)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Figure 1C"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1177dda90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the probabilty distribution of classified mutations\n",
    "from numpy import inf\n",
    "bins = pd.cut(downsampled.Y_probabilities['True'], 20, retbins=True)[0]\n",
    "bins = pd.DataFrame(bins)['True'].value_counts(sort=False)\n",
    "bins[bins == -inf] = 0\n",
    "bins.index = np.array(range(0,100,5))/100.  ##\n",
    "plt.figure(figsize=(16,12))\n",
    "sns.set_context(\"talk\", font_scale=1.8)\n",
    "sns.set_style(\"white\")\n",
    "ax = sns.barplot(\n",
    "                    list(bins.index), \n",
    "                    list(bins), \n",
    "                    palette=options.palette,\n",
    "                    edgecolor='#ffffff', linewidth=5\n",
    "                )\n",
    "for patch in ax.patches:\n",
    "    patch.set_edgecolor('#ffffff')\n",
    "    patch.set_linewidth(2)\n",
    "    patch.set_alpha(0.85)\n",
    "\n",
    "ax.set(xlabel='Driver Probability', ylabel='Count')\n",
    "ax.set_xticklabels(ax.get_xticklabels(), rotation = 45)\n",
    "n = 2  # Keeps every other\n",
    "[l.set_visible(False) for (i,l) in enumerate(ax.xaxis.get_ticklabels()) if i % n != 0]\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading prostate model...\n",
      "CPU times: user 966 ms, sys: 188 ms, total: 1.15 s\n",
      "Wall time: 1.23 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Now create a model using the full downsampled data (without cross validation)\n",
    "if options.use_saved and os.path.exists(\"prostate23_model.pkl\"):\n",
    "    print \"Loading prostate model...\"\n",
    "    full_model = load_matrix(\"prostate23_model.pkl\")\n",
    "else:\n",
    "    print \"Generating prostate model...\"\n",
    "    #full_model = MutationMatrix(db_uri=options.prostate_dbi)\n",
    "    full_model = downsampled.copy()\n",
    "    full_model.set_features(new_features)\n",
    "    full_model.svm(cross_validate=False, test_size=0.0, class_weight=None)\n",
    "    full_model.save_matrix(\"prostate23_model.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading prostate final scores...\n",
      "CPU times: user 1min 58s, sys: 14.3 s, total: 2min 12s\n",
      "Wall time: 2min 22s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Apply the model to all mutations\n",
    "if options.use_saved and os.path.exists(\"prostate23_final.pkl\"):\n",
    "    print \"Loading prostate final scores...\"\n",
    "    all_mutations = load_matrix(\"prostate23_final.pkl\")\n",
    "else:\n",
    "    print \"Generating prostate final scores...\"\n",
    "    #all_mutations = MutationMatrix(db_uri=options.prostate_dbi)\n",
    "    all_mutations = mutations.copy()\n",
    "    # Use the same selected features\n",
    "    all_mutations.set_features(list(full_model.features))\n",
    "    # And normalize the same way\n",
    "    X = full_model._normalize(all_mutations[full_model.features], \n",
    "                              imputer=full_model.imputer, \n",
    "                              scaler=full_model.scaler, \n",
    "                              as_df=True)[0]\n",
    "    # Determine distances from the discriminatory hyperplane\n",
    "    all_mutations['Distance'] = full_model.model.decision_function(X)\n",
    "    all_mutations = all_mutations.sort_values(by='Distance')    \n",
    "    all_mutations.save_matrix('prostate23_final.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lone Feature Predictor Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Repeat for dann, phlyloP, cadd, cadd_raw, funseq2, and aa_class_change|unchanged\n",
    "fws = pd.DataFrame(np.nan, index=new_features, columns=['Weight'])\n",
    "fws.index.name='Feature'\n",
    "fws = fws.reset_index()\n",
    "for fea in new_features:\n",
    "    print \"Evaluating\", fea\n",
    "    feature_test = mutations.copy()\n",
    "    feature_test.normalize_options['scaler_strat']='mms'\n",
    "    feature_test.normalize_options['nan_strat']='mean'\n",
    "    feature_test.set_features([fea])\n",
    "    feature_test = feature_test.dropna(subset=[fea])\n",
    "    feature_test.svm(cross_validate=False, test_size=0.0)\n",
    "    ff  = feature_test.show_feature_importances('table')\n",
    "    fws.loc[fws['Feature']==fea,'Weight'] \\\n",
    "      = feature_test.show_feature_importances('table').ix['False',fea]\n",
    "fws = fws.sort_values('Weight', ascending=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fws"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# And plot them\n",
    "span, bot, top = (20,0,20)\n",
    "mod = sns.color_palette(cmap(map(lambda k: k/float(top-bot), range(0,top-bot))))\n",
    "\n",
    "sns.set_context(\"poster\")\n",
    "sns.set_style(\"white\")\n",
    "fig, ax = plt.subplots(figsize=(11.7, 8.27))\n",
    "\n",
    "ax=sns.barplot(y='Feature', x='Weight', data=fws,\n",
    "                linewidth=2, palette=mod)\n",
    "#ax.set_xlim(-3,3)\n",
    "\n",
    "\n",
    "for i,artist in enumerate(ax.artists):\n",
    "    # Set the linecolor on the artist to the facecolor, and set the facecolor to None\n",
    "    col = artist.get_facecolor()\n",
    "    artist.set_edgecolor(col)\n",
    "    artist.set_facecolor('None')\n",
    "    artist.set_zorder(10)\n",
    "\n",
    "    # Each box has 6 associated Line2D objects (to make the whiskers, fliers, etc.)\n",
    "    # Loop over them here, and use the same colour as above\n",
    "    for j in range(i*6,i*6+6):\n",
    "        line = ax2.lines[j]\n",
    "        line.set_color(col)\n",
    "        line.set_mfc(col)\n",
    "        line.set_mec(col)\n",
    "\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Full Model Feature Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    }
   },
   "outputs": [],
   "source": [
    "### FOR ORIGINAL WEIGHTS!\n",
    "# Get feature weights \n",
    "fws = full_model.show_feature_importances('table')\n",
    "fws = fws.ix['False',new_features] ## To make passenger enriched on the right side of plot\n",
    "fws = fws.transpose().reset_index()\n",
    "fws.columns = ['Feature', 'Weight']\n",
    "fws = fws.sort_values('Weight', ascending=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Supplemental Figure 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "run_control": {
     "frozen": false,
     "read_only": false
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x13482f810>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# And plot them\n",
    "span, bot, top = (20,0,20)\n",
    "mod = sns.color_palette(cmap(map(lambda k: k/float(top-bot), range(0,top-bot))))\n",
    "\n",
    "sns.set_context(\"poster\")\n",
    "sns.set_style(\"white\")\n",
    "fig, ax = plt.subplots(figsize=(11.7, 8.27))\n",
    "\n",
    "ax=sns.barplot(y='Feature', x='Weight', data=fws,\n",
    "                linewidth=2, palette=mod)\n",
    "#ax.set_xlim(-3,3)\n",
    "\n",
    "\n",
    "for i,artist in enumerate(ax.artists):\n",
    "    # Set the linecolor on the artist to the facecolor, and set the facecolor to None\n",
    "    col = artist.get_facecolor()\n",
    "    artist.set_edgecolor(col)\n",
    "    artist.set_facecolor('None')\n",
    "    artist.set_zorder(10)\n",
    "\n",
    "    # Each box has 6 associated Line2D objects (to make the whiskers, fliers, etc.)\n",
    "    # Loop over them here, and use the same colour as above\n",
    "    for j in range(i*6,i*6+6):\n",
    "        line = ax2.lines[j]\n",
    "        line.set_color(col)\n",
    "        line.set_mfc(col)\n",
    "        line.set_mec(col)\n",
    "\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}